Split gate non-volatile flash memory cells are well known. For example, U.S. Pat. No. 6,747,310 discloses such memory cells having source and drain regions defining a channel region there between, a select gate over one portion of the channel regions, a floating gate over the other portion of the channel region, and an erase gate over the source region. The memory cells are formed in pairs that share a common source region and common erase gate, with each memory cell having its own channel region in the substrate extending between the source and drain regions (i.e. there are two separate channel regions for each pair of memory cells). The lines connecting all the control gates for memory cells in a given column run vertically. The same is true for the lines connecting the erase gates and the select gates, and the source lines. The bit lines connecting drain regions for each row of memory cells run horizontally.
Each memory cell stores a single bit of information (based on the programming state of the floating gate). Given the number of electrodes for each cell (source, drain, select gate, control gate and erase gate), and two separate channel regions for each pair of memory calls, configuring and forming the architecture and array layout with all the various lines connected to these electrodes can be overly complex and difficult to implement, especially as critical dimensions continue to shrink.
One solution is to eliminate the source region, and have both memory cells share a single continuous channel region and a common word line gate, and disclosed in U.S. Pat. No. 8,780,625. However, there are performance limitations with this configuration because, among other things, it lacks erase gates.